Uploaded model
- Developed by: Yammer123
- License: apache-2.0
- Finetuned from model : llm-jp/llm-jp-3-13b
- **Dataset：_ichikara-instruction, _databricks-dolly-15k-ja, tv_information
This llama model was trained 2x faster with Unsloth and Huggingface's TRL library.
Python
Google Colab の場合は上記の環境構築手順を行なわず、単にこのセルから実行していってください。
!pip uninstall unsloth -y
!pip install --upgrade --no-cache-dir "unsloth[colab-new] @ git+
https://github.com/unslothai/unsloth.git"
!pip install --upgrade torch
!pip install --upgrade xformers
!pip install ipywidgets --upgrade
Install Flash Attention 2 for softcapping support
import torch
if torch.cuda.get_device_capability()[0] >= 8:
!pip install --no-deps packaging ninja einops "flash-attn>=2.6.3"
llm-jp/llm-jp-3-13bを4bit量子化のqLoRA設定でロード。
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
from unsloth import FastLanguageModel
import torch
max_seq_length = 512 # unslothではRoPEをサポートしているのでコンテキスト長は自由に設定可能
dtype = None # Noneにしておけば自動で設定
load_in_4bit = True # 今回は8Bクラスのモデルを扱うためTrue
model_id = "Yammer123/llm-jp-3-13b-finetune-2"
FastLanguageModel インスタンスを作成
model, tokenizer = FastLanguageModel.from_pretrained(
model_name=model_id,
dtype=dtype,
load_in_4bit=load_in_4bit,
trust_remote_code=True,
)
SFT用のモデルを用意
model = FastLanguageModel.get_peft_model(
model,
r = 64,#32
target_modules = ["q_proj", "k_proj", "v_proj", "o_proj",
"gate_proj", "up_proj", "down_proj",],
lora_alpha = 64,#32
lora_dropout = 0.075,
bias = "none",
use_gradient_checkpointing = "unsloth",
random_state = 3407,
use_rslora = False,
loftq_config = None,
max_seq_length = max_seq_length,
)
HF_TOKEN = "YOUR KEY"
ELYZA-tasks-100-TVの読み込み。事前にファイルをアップロードしてください
import json
datasets = []
with open("./elyza-tasks-100-TV_0.jsonl", "r") as f:
item = ""
for line in f:
line = line.strip()
item += line
if item.endswith("}"):
datasets.append(json.loads(item))
item = ""
学習したモデルを用いてタスクを実行
from tqdm import tqdm
推論するためにモデルのモードを変更
FastLanguageModel.for_inference(model)
results = []
for dt in tqdm(datasets):
input = dt["input"]
prompt = f"""### 指示\n{input}\n### 回答\n"""
inputs = tokenizer([prompt], return_tensors = "pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens = 512, use_cache = True, do_sample=False, repetition_penalty=1.2)
prediction = tokenizer.decode(outputs[0], skip_special_tokens=True).split('\n### 回答')[-1]
results.append({"task_id": dt["task_id"], "input": input, "output": prediction})
jsonlで保存
with open(f"./_output.jsonl", 'w', encoding='utf-8') as f:
for result in results:
json.dump(result, f, ensure_ascii=False)
f.write('\n')
